POST Diagnostics Card

[Guest]

PCI Post Diagnostics Test Card
StarTech.com ID: PCIPOST

I suspected a dead mainboard or chip since the CPU fan does not turn
and the screen is blank. Yet with this POST diags card Code 42
displayed refers to hard disk drive and controller. The hard drive
works fine in another PC. The mainboard/chip/RAM/vga should all give a
screen display irrespective of whether the hard drives even exist. My
bios is Award and the card for this manual identifies code 42 as a hard
drive error, even with no hard drive present!
Is this POST card any good or could someone suggest another?

BJ

[Charles Howse]

Previously fastrack1966@clara.co.uk wrote: PCI Post Diagnostics Test Card StarTech.com ID: PCIPOST
I suspected a dead mainboard or chip since the CPU fan does not turn and the screen is blank. Yet with this POST diags card Code 42 displayed refers to hard disk drive and controller. The hard drive works fine in another PC. The mainboard/chip/RAM/vga should all give a screen display irrespective of whether the hard drives even exist. My bios is Award and the card for this manual identifies code 42 as a hard drive error, even with no hard drive present! Is this POST card any good or could someone suggest another?

The card does only display what the BIOS tells it to by writing
to a specific port. Changing the card will not change the code.

As for the HDD-error, maybe the HDD scan (i.e. trying to find
whwther there is a HDD) could be the problem. This could be a
fried controller.

Arno

[Guest]

Thanks Arno that makes sense. I suppose I assumed the board and cpu
would display something on the monitor then look at hard
drives/controllers nanoseconds later, eg at the "Detecting IDE Drives"
message.
My super new POST card is now displaying FF on a different board one
with an AMI bios, but FF does not show in the AMI codes list. Does
anyone have any ideas on the one please?

[Charles Howse]

Previously fastrack1966@clara.co.uk wrote: Thanks Arno that makes sense. I suppose I assumed the board and cpu would display something on the monitor then look at hard drives/controllers nanoseconds later, eg at the "Detecting IDE Drives" message. My super new POST card is now displaying FF on a different board one with an AMI bios, but FF does not show in the AMI codes list. Does anyone have any ideas on the one please?

"ff" means "BIOS operation finished". The POST-codes always increase
when some stage has been reached. 0xff = 256 ist the largest (=last)
value.

Arno

[Guest]

Thanks Arno.
Sorry to labour this point everyone but the "manual" is a fat lot of
good. Clarifying wnat the author means is hard when a manufacturer in
the Far East has not bothered to hire a proper interpreter to write the
manual - we've all seen this kind of work!
The book says if a test fails "the system will pause and the error code
will keep showing on the card"
Can I therefore assume that:
1) Other errors may lie ahead, and will not be revealed until I've
fixed the first displayed error.
2) If the card reaches FF without pausing all is well?
Also the card has two lights that are mentioned nowhere in the booklet:
one called reset pass/error and the other clk pass/error. With my error
42 the reset light is on (pass) but the clk light is off (error).
Obviously both lights are meant to be on for success, but what are we
meant to make of only one light being on?

Again many thanks for helping me read this booklet!

[Guest]

Sorry I should have added one thing - I have a mobo here that is going
through to FF but only the top (RESET) light is on. The lower (CLK)
light is off. Whoops, maybe FF is Finally Finished not Feeling Fine?

[B Frank]

If your Startech card is working properly you would see
various hexadecimal codes of varying duration flash by before
the Power On Self Test Card stops at "42" or "FF". Otherwise
you are observing the power-on state of the diagnostic card
or the PCI bus. Thus "FF" and "42" would be meaningless
codes.
Frankly, the Startech POST card (which is actually made by
Sunix, their model 9501) is a piece of junk that does not
work properly like other Startech products. I presume you're
talking about http://www.startech.com . Startech.com is
(or was) just an importer of the most inexpensive Chinese
"stuff" sold under their house brands. Startech claims to be
a manufacturer, but they are NOT. We went through a few of
the Startech PCI POST diagnostic cards, their NICs and other
Startech.com products before encountering more compatibility
issues than the exercise was worth. Here is the problem with
the Startech PCI debug card:

When booting, the BIOS checks the PCI slots for installed
PCI cards. If no PCI card is detected the BIOS assumes
it has an empty PCI slot and will turn off the PCI-clock
signal to the empty slots. PCI slots with a POST diagnostic
card plugged in look like empty PCI slots to the BIOS
(they have to be, otherwise the presence of POST cards
would alter the test, since the BIOS would have to allocate
resources to it). POST diagnostics cards are passive devices
that intercept POST error codes broadcast blindly by the
BIOS to the bus. This PCI-clock "off" feature can be disabled
in the AMI, Award etc. BIOS setups, but PCI-clock "off" to
unused slots is the BIOS default. There is no way to turn
PCI-clock on once the mobo fails.

The better and more expensive POST diagnostic cards work
without the PCI-clock bus signal. These cards have circuitry
to overcome the absence of the PCI-Clock and also have
a switch to select from different POST ports (like port 84h
for Compaq machines). Select the wrong POST port or intercept
8-bit codes from more than one port at-a-time and the codes
you see are anything but POST error codes.

Note: JDR, Startech and others resell the same Chinese
PCI diagnostics card as per:
<http://www.sunix.com.tw/en/Promotion-1.php?sid=18>
<http://64.233.167.104/search?q=cache:rQEAugWcYSEJ:https://startech.com/ProductSpecs/PCIPOST.htm+post+card+pci+site:startech.com&hl=en>
<http://jdr.com/interact/item.asp?itemno=gr-pcode-p>
Years ago we sent our PCI POST cards back to Startech
for a refund! I am surprised that this junk is still
being sold.

BTW, POST diagnostics cards work quite well as diagnostic tools.
Negative comments about POST cards started to appear when the
more expensive locally made product was displaced by buggy
inexpensive Asian copies with useless manuals. In the
absence of a good manual you may want to look at "The BIOS
Companion" book by Phil Croucher, available at amazon.com
or at the author's web site http://www.electrocution.com
His book has well over 100 pages of POST error codes with
explanations besides a wealth of other hard to find information.

For what it is worth, here are some archived google groups
searches on this topic. Have a look at the threads and messages.
<http://groups.google.com/groups?as_q=&num=100&scoring=d&hl=en&as_epq=diagnostic+cards&as_oq=&as_eq=&as_ugroup=&as_usubject=&as_uauthors=&lr=&as_drrb=q&as_qdr=&as_mind=1&as_minm=1&as_miny=1981&as_maxd=11&as_maxm=6&as_maxy=2005&safe=off>
<http://groups.google.com/groups?as_q=pci&num=100&scoring=r&hl=en&as_epq=post+card&as_oq=&as_eq=jpmcbooks&as_ugroup=&as_usubject=&as_uauthors=&lr=&as_drrb=q&as_qdr=&as_mind=1&as_minm=1&as_miny=1981&as_maxd=11&as_maxm=6&as_maxy=2005&safe=off>
<http://groups.google.com/groups?as_q=&num=100&scoring=d&hl=en&as_epq=debug+card&as_oq=&as_eq=&as_ugroup=&as_usubject=&as_uauthors=&lr=&as_drrb=q&as_qdr=&as_mind=1&as_minm=1&as_miny=1981&as_maxd=11&as_maxm=6&as_maxy=2005&safe=off>
<http://groups.google.com/groups?as_q=&num=100&scoring=d&hl=en&as_epq=diagnostic+card&as_oq=&as_eq=&as_ugroup=&as_usubject=&as_uauthors=&lr=&as_drrb=q&as_qdr=&as_mind=1&as_minm=1&as_miny=1981&as_maxd=11&as_maxm=6&as_maxy=2005&safe=off>
<http://groups.google.com/groups?as_q=&num=100&scoring=d&hl=en&as_epq=diagnostics+card&as_oq=&as_eq=&as_ugroup=&as_usubject=&as_uauthors=&lr=&as_drrb=q&as_qdr=&as_mind=1&as_minm=1&as_miny=1981&as_maxd=11&as_maxm=6&as_maxy=2005&safe=off>

Hope this helps.

Copyright (c) 2005 by fcwilson at canoemail dot com


fastrack1966@clara.co.uk wrote: Thanks Arno. Sorry to labour this point everyone but the "manual" is a fat lot of good. Clarifying wnat the author means is hard when a manufacturer in the Far East has not bothered to hire a proper interpreter to write the manual - we've all seen this kind of work! The book says if a test fails "the system will pause and the error code will keep showing on the card" Can I therefore assume that: 1) Other errors may lie ahead, and will not be revealed until I've fixed the first displayed error. 2) If the card reaches FF without pausing all is well? Also the card has two lights that are mentioned nowhere in the booklet: one called reset pass/error and the other clk pass/error. With my error 42 the reset light is on (pass) but the clk light is off (error). Obviously both lights are meant to be on for success, but what are we meant to make of only one light being on? Again many thanks for helping me read this booklet!

[Bob Eager]

On Sat, 11 Jun 2005 15:51:23 UTC, fastrack1966@clara.co.uk wrote:
Sorry I should have added one thing - I have a mobo here that is going through to FF but only the top (RESET) light is on. The lower (CLK) light is off. Whoops, maybe FF is Finally Finished not Feeling Fine?

Since all these numbers are BIOS dependent, you really need to refer
only to actual BIOS documentation.

This may help (at a small cost):

http://www.electrocution.com/biosc.htm

--
Bob Eager
begin a new life...take up Extreme Ironing!

[B Frank]

fastrack1966@clara.co.uk wrote: Sorry I should have added one thing - I have a mobo here that is going through to FF but only the top (RESET) light is on. The lower (CLK) light is off. Whoops, maybe FF is Finally Finished not Feeling Fine?

"FF" hexadecimal is "1111 1111" in binary or the highest
of the 256 possible POST codes. Normally "FF" is the
power-on default state of the POST card's display register.

Clock LED = "OFF" - meaning is that there is NO PCI-Clock
present, so the POST code you see is meaningless (unless
the POST card has circuitry to overcome the PCI-clock problem).
Plug the POST card into a working system and look for a BIOS
setting in one of the sub-menus with the words "DIMM/PCI Clk"
or "Clock for empty PCI Slot/DIMM" etc. - change them. Now
the Clock LED on your diagnostic card should come "ON" and
the codes displayed are the actual POST codes as the BIOS
progresses through the POST sequence.

RESET LED = "ON" - normally means that the RESET line from
the PSU or the RESET switch is stuck. Mobo is in a permanent
Reset state, thus nothing happens, again the POST code you
see is meaningless. Note: on some Diagnostic cards the Reset
LED works inverted, so LED = "OFF" means stuck Reset line.

Note: POST codes DO NOT appear in order during the boot
process. The boot sequence changes from BIOS to BIOS
and on BIOS features selected. This is quite normal.

Copyright (c) 2005 by fcwilson at canoemail dot com

[Jason Gurtz]

On 6/11/2005 12:43, B Frank wrote:
Frankly, the Startech POST card (which is actually made by Sunix, their model 9501) is a piece of junk that does not work properly like other Startech products.

FWIW, I've been using startech.com PCI LPT port cards (1 and 2-port) under
Win2k and 2k3 and they've worked surprisingly well. Only other stuff I
get from them is cables.

cheers,

~Jason

--

[FredBear]

Thanks everyone. Some interesting material for debate there so what
I've done is:
1) Returned the card as unfit for its purpose
2) Stacked away all my redundant mobos for another day!
If anyone can suggest an alternative brand of diagnostic card that has
greater regard from its users I'd be grateful

BJ

[Guest]

Fastrack wrote:
<snip>
If anyone can suggest an alternative brand of diagnostic card that has greater regard from its users I'd be grateful BJ

I was recommended the following ones. though they're pricy.

$700 excellent one
http://www.uxd.com/phdpci2.shtml
Apparently, on a quad Xeon system, with four $900 Xeon chips, it's nice
to know what's wrong


a lot less good one $300
http://www.protechdiagnostics.com/pccertify_prokit.htm


this is one mentioned by scott mueller. $200
PC Certify PCISA flippost card